I'll start a thread on this topic and will take up Ians' suggestion of using a 6Y9 Dekal base valve and his suggestion of the arrangement.

The smaller pentode should work nicely as an Autodyne and I can reflex the output as Ian suggests for RF and AF.
The front end will be like a pentode autodyne, usually done with a type 57 in the old days.

The AF output section will be a reflex action with the plate circuit working almost as Triode to drive the speaker with AF.
The screen grid will be the RF "plate" of a triode to loop out the IF signal and feedback to the control grid.
So the pentode will sort of be like a twin triode, with a common control grid with IF and AF stuck into it.
What could possibly go wrong!!!!!

I have a circuit in my head, and have some 6Y9 in a junk box somewhere. Might have to make up a socket to hold the valve.
I can use some mini IFT, a loopstick aerial coil, wind up an osc coil, get a small gang off something, and wind up a little power tranny and choke for the HT. Should be a doddle.

Any thoughts or suggestions always welcome, fire away.

Fred.

6Y9 Superhet - Part 1
6Y9 Superhet - Part 2
6Y9 Superhet - Part 3
6Y9 Superhet - Part 4
Kriesler 11-99 Rebuild

Great project Fred.
Silicon rectifier diodes for power supply and then a 0ne valve superhet.
10 pin bases available Ebay for $3, ceramic ones at that.
JJ

Hi Johnny, yep ordered some bases from that guy on ebay.
Will start some prelim work, gather up the coils, power supply and support stuff.
Fred.

Fred, if I get a chance this weekend I'll draw up a circuit as I visualise it. I'm thinking it'll have AGC and negative feedback!

The 6Y9 as you may know was designed to handle higher frequency stuff than we are asking of it. The main issue I can foresee is keeping it stable.

Very, very busy at work, we are an engineer down and shortly to be two down. The young guy who was at the next desk took a job at NASA designing inertial navigation systems for the moon rovers.

How cool is that! Every young engineer's dream job!

I like the story of the Russian one: Great giggle as it fell into a crater and the west thought that is where it went. Of course the Russians got the last laugh as it actually found its way out.

Ian, thanks for any thoughts.
I plan to protoype a simple version with the mixer 1/2 driving one IFT then a diode and straight into the output audio 1/2.
That will allow me to get my tuning coils and IF details sorted out without any stability issues.
I can then add in the reflexing and sort the gain out as I go.
The 6Y9 looks like a pretty "hot" valve so it will be interesting to see how little I have to do to get performance.

Fred.

I've drawn up a preliminary schematic and I'm about to send Brad a copy for posting.

What it does not yet have is some means of adjusting the RF gain. Ideas are requested!

You can't use normal AGC biassing techniques on either the converter or the IF, since it is also the output stage.

A neon and an LDR?

Click on image for larger resolution

I did an LTSpice simulation of a back to back diode attenuator using two ordinary 1N4148 diodes between the secondary of IFT1 and the grid of the following stage.

By deriving a negative voltage in the same way AGC is normally derived, connecting this to the two anodes via a 47k, pulling it up ever so lightly with a 2.7 meg and with a 150 millihenry choke on the output end, I can get about 30db of fairly linear control with a control voltage varying from 0 to -4 volts.

The untuned choke worked best, tuning it gave more gain but greatly reduced the amount of attenuation possible. As it is, a very strong signal pretty much cuts it off. Just like "real" AGC!

I'll see if I can get a screen grab put up.

The green trace is the control voltage, the blue is the 455kHz.

I think we are allowed to use silicon diodes, they are not "active" devices. Does this break any rules?

Hi Ian, silicon diodes are fine, in keeping with that era.
I could use germanium for the detector/agc and I have a selenium bridge for the HT!
Only kidding, silicon all the way on this one.
Rules?
We make the rules here!
In the first version I wont bother at all with AGC, I will mainly work on winding the coils and matching them to the gang so I can get a decent 455kc signal out of the mixer/osc on the AM 550/1800 kc range.

I'm beavering away on the mechanical modulator at the moment.
I want to get that to a useable state first.
I can only stand up and work in the workshop for a couple hours on any day.
This "long covid" is not funny.
At least I stayed out of hospital but this is 5 months now and still affecting me.

Fred.

Photos uploaded to Posts 7 and 8.

What it does not yet have is some means of adjusting the RF gain. Ideas are requested!

Do it the old-fashioned way. Have a pot to progressively short the aerial input to ground.

Re post #7 and circuit.

A/ Thanks Ian, your and my circuit are hellishly close!

One major difference is I have proposed to reflex out of the screen grid of the output rather than the plate.
I am not sure of the current handling capability of the mini IFT's I might use.
I can always use standard valve IFT's and follow your plate circuit.

Your way the output valve is a full pentode and not 2 psuedo triodes.
There is a lot of theory about the 2 methods in the AWV notes and my Langford-Smith radio handbook I can bone up on.

B/ RobB ,Yep thats another thing to consider as well, thanks.

Fred.

Photo uploaded to Post 7. Embedded image is a link to the full size image this time, so it can be seen more clearly.

Thanks Brad

What is the best resolution to use when sending screen grabs like these? Single pixel lines don't scale too well.

All photos are currently shrunk to 900px wide. It is okay to make them larger if clarity is important. It's no problem to add a link to a larger image.